memory - rbvtaylor.weebly.com · atkinson & shiffrin model of memory ... w o r k s b o b d a e...

MEMORY

ATKINSON & SHIFFRIN MODEL OF MEMORY

Consists of three stores or functional storage location Sensory store Short-term store Long-term store

ENCODING, STORAGE AND RETRIEVAL

The three stores each use three processes:1. Encoding, or the process of storage, 2. Moving of the encoded memory into a memory

store and maintaining that information3. Retrieval, which is the subsequent access of the

stored memory information in order to bring it back into consciousness

103

STUDYING MEMORY: INFORMATION PROCESSING MODELS

Keyboard(Encoding)

Disk(Storage)

Monitor(Retrieval)

Sequential Process

104

INFORMATION PROCESSINGThe Atkinson-Schiffrin (1968) three-stage model

of memory includes a) sensory memory, b)short-term memory, and c) long-term memory.Bo

b D

aem

mric

h/ T

he Im

age

Wor

ks

Bob

Dae

mm

rich/

The

Imag

e W

orks

Fran

k W

arte

nber

g/ P

ictu

re P

ress

/Co

rbis

Atkinson &Schiffrin

MULTIPLE-STORE MODEL OF MEMORY

SENSORY MEMORY

The iconic store is a visual memory that lasts for only a fraction of a second

Sperling’s research demonstrated that subjects could recall visual material (icons), but only within a very short (less than a second) time interval. George Sperling

107

WHOLE REPORT

The exposure time for the stimulus is so smallthat items cannot be rehearsed.

R G TF M QL Z S

50 ms (1/20 second)

“Recall”R T M Z

(44% recall)

Sperling (1960)

108

PARTIAL REPORT

Low Tone

Medium Tone

High Tone

“Recall”J R S

(100% recall)

Sperling (1960) argued that sensory memory capacity was larger than what was originally thought.

50 ms (1/20 second)

S X TJ R SP K Y

109

TIME DELAY

“Recall”N _ _

(33% recall)

TimeDelay

50 ms (1/20 second)

A D IN L VO G H

Low Tone

Medium Tone

High Tone

110

SENSORY MEMORY

The longer the delay, the greater the memory loss.

20

40

60

80

Perc

ent R

ecog

nize

d

0.15 0.30 0.50 1.00

Time (Seconds)

111

SENSORY MEMORIES

Iconic0.5 sec. long

Echoic3-4 sec. long

Hepatic< 1 sec. long

The duration of sensory memory varies for the different senses.

ICONIC MEMORY AND MOTION PICTURES

Why do we see movement, when in reality, what we are seeing is a rapidly presented series of still pictures?

SENSORY STORAGE

An echoic store also exists for the storage of perceived sounds, as well as possible sensory stores for the other sensory modalities (such as olfaction and taste – but these latter two are controversial)

SHORT-TERM MEMORY

A storage space of limited capacity and duration

At most, information can be held in the short term store for up to a couple of minutes

Typically held for only a few seconds

115

CAPACITY

You should be able torecall 7±2 letters.

The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for

Processing Information (1956).

George Miller

M U T G I K T L R S Y P

Ready?

116

CHUNKING

F-B-I-T-W-A-C-I-A-I-B-M

The capacity of the working memory may be increased by “chunking.”

FBI TWA CIA IBM4 chunks

117

ENCODING: GETTING INFORMATION IN

How We Encode

1. Some information (route to your school) is automatically processed.

2. However, new or unusual information (friend’s new cell-phone number) requires attention and effort.

118

AUTOMATIC PROCESSINGWe process an enormous amount of information

effortlessly, such as the following:

1. Space: While reading a textbook, you automatically encode the place of a picture on a page.

2. Time: We unintentionally note the events that take place in a day.

3. Frequency: You effortlessly keep track of things that happen to you.

119

EFFORTFUL PROCESSING

Committing novel information to memory requires effort just like learning a concept from

a textbook. Such processing leads to

durable and accessible memories.

Spencer Grant/ Photo Edit

© Bananastock/ Alam

y

ENCODING

Studying how information is encoded in the short-term store, Conrad found that strings of letters (e.g., B, C, F, M, etc.) presented visually were most subject to acoustic confusability (F vs. S).

Baddeley produce similar findings in an experiment that varied the acoustic and semantic similarities of similar words.

Suggests short-term memory encoded acoustically, not semantically.

REHEARSAL

Once information is in the short-term store, how is it kept there?

Rehearsal (in this case, maintenance rehearsal of repeating the information over and over) is the key to keeping items in short-term memory.

Strategies like rehearsal are examples of meta-memory skills, or skills based on an understanding of how one’s own mind works.

122

REHEARSAL

Effortful learning usually requires

rehearsal or conscious repetition.

Ebbinghaus studied rehearsal by using nonsense syllables:

TUV YOF GEK XOZ

Hermann Ebbinghaus(1850-1909)

http://ww

w.isbn3-540-21358-9.de

123

REHEARSAL

The more times the nonsense syllables were

practiced on Day 1,the fewer repetitions were

required to remember them on Day 2.

REHEARSAL

Mnemonic devices are also examples of meta-memory skills.

Young children, typically, have not yet learned metamemory or other metacognitive skills.

INTERFERENCE AND DECAY

When information is lost from short-term memory, forgetting occurs.

The two best studied theories on forgetting are interference and decay theory.

INTERFERENCE

Discovered simultaneously by researchers Brown and Peterson when they found that sometimes subject’s forgetting was due to the fact that the target information was displaced in memory by competing information.

The two types of interference that have been identified:Retroactive interference Proactive interference

RETROACTIVE INTERFERENCE

Occurs when newer information in memory inhibits the retrieval of older information in memory. (Ex.: one has trouble remembering one’s previous phone number because it is blocked by the newly memorized new phone number)

PROACTIVE INTERFERENCE

Occurs when older information in memory inhibits the retrieval of newer information in memory. (Ex.: continually calling a new roommate by the name of a previous roommate)

DECAY THEORY

Memories disappear gradually with disuse. Difficult to test for because experimenters can

never be sure that items have not recently been rehearsed.

CAPACITY OF SHORT-TERM MEMORY

Miller demonstrated that short-term memory had capacity of around seven items, plus or minus two items.

Items can be chunked together so seven words can be remembered or seven letters if they did not make up a word.

CHUNKING

134

CHUNKING

Organizing items into a familiar, manageable unit. Try to remember the numbers below.

1-7-7-6-1-4-9-2-1-8-1-2-1-9-4-1

If you are well versed with American history, chunk the numbers together and see if you can recall them better. 1776 1492 1812 1941.

135

CHUNKING

Acronyms are another way of chunking information to remember it.

HOMES = Huron, Ontario, Michigan, Erie, Superior

ROY G. BIV = Red, Orange, Yellow, Green, Blue, Indigo, Violet

CAPACITY OF SHORT-TERM MEMORY

The serial position curve shows how the order of presentation affects short-term memory.

Items near the middle of any set of items are the hardest to remember.

Items to the front of the set are more easily remembered, a phenomenon known as the primacy effect. The primacy effect probably occurs because the first words have been moved into long-term memory by the time they are recalled.

CAPACITY OF SHORT-TERM MEMORY

Items near the end of the set also were easily remembered, a phenomenon known as the recency effect. Recency probably occurs because these words are still held in short-term memory at the time of recall.

SHORT-TERM MEMORY RETRIEVAL

Saul Sternberg’s research in memory scanning subjects given a string of one to six digits to hold short-term memory. After a short pause, a test digit flashed before them on the screen.

Possible explanations for retrieval: Parallel processing Serial processing

ExhaustiveSelf terminating

PARALLEL PROCESSING

Perhaps the subjects were able to simultaneously retrieve all the digits from short-term memory and compare them to the test digit. If so, the size of the original string of digits should not impact response time.

SERIAL PROCESSING

Perhaps the subjects sequentially retrieved the digits from short-term memory and compared them to the test digit. If so, then the size of the original digit string would influence the response time.

141

LONG-TERM MEMORY

Essentially unlimited capacity store.

The Clark’s nutcracker can locate 6,000 caches ofburied pine seeds during winter and spring.

R.J. Erwin/ Photo Researchers

142

MEMORY FEATS

LONG-TERM MEMORY

Long-term memories are what people usually are referring to speak of memory.

May last indefinitely, but how are they moved into long-term memory?Rehearsal Paying deliberate attention to information Connecting new information to old Integrating new information into memory

FORMS OF ENCODING IN LONG-TERM MEMORY

Unlike short-term memory, long-term memory is mostly encoded semantically, or via meaning.

But visual and acoustic encoding are also present

Blousfield found that subjects are more likely to retrieve words by category than at random

Frost demonstrated that subjects may simultaneously use both semantic and visual information when encoding.

FORMS OF ENCODING IN LONG-TERM MEMORY

Nelson and Rothbart suggest that the acoustic codes can operate in long-term memory as well as in short-term memory

Anderson and Bower and others have produced data that support the propositional view (storage according to their deep meaning) for storage of both verbal and visual information.

MacLeod, Hunt, and Matthews found data that suggest the subjects use either propositional or imaginal storage.

146

STORING MEMORIES IN THE BRAIN

1. Loftus and Loftus (1980) reviewed previous research data showing, through brain stimulation, that memories were etched into the brain and found that only a handful of brain stimulated patients reported flashbacks.

2. Using rats, Lashley (1950) suggested that even after removing parts of the brain, the animals retain partial memory of the maze.

147

SYNAPTIC CHANGES

In Aplysia, Kandel and Schwartz (1982) showed that serotonin release from neurons increased

after conditioning.

Photo: Scientific American

148

SYNAPTIC CHANGES

Long-Term Potentiation (LTP) refers to synaptic

enhancement after learning (Lynch, 2002).

An increase in neurotransmitter release

or receptors on the receiving neuron

indicates strengthening of synapses.

Both Photos: From N

. Toni et al., Nature, 402, N

ov. 25 1999. Courtesy of Dom

inique Muller

BADDELEY’S MODEL: WORKING MEMORY

Working memory is seen as a special, activated part of long-term memory that moves information into and out of short-term memory.

Working memory consists of three parts: Visuospatial sketchpad – briefly holds visual

information Phonological loop – briefly holds auditory

information Central executive – coordinates attentional

activities and governance responses.

BADDELEY’S MODEL: WORKING MEMORY

Recently, another component has been added to the model, the episodic buffer. The job of this limited-capacity component is to take information from the different parts of working memory so that they make sense

BADDELEY’S MODEL: WORKING MEMORY

Some physiological data support the working memory view of memory. PET data have yielded preliminary mappings of the brain areas that are related to the visiospatial sketchpad, phonological loop, and central executive. Further research is necessary to confirm validity of the working memory theory.

BADDELEY’S MODEL: WORKING MEMORY

Some physiological data support the working memory view of memory. PET data have yielded preliminary mappings of the brain areas that are related to the visiospatial sketchpad, phonological loop, and central executive. Further research is necessary to confirm validity of the working memory theory.

153

STORING IMPLICIT & EXPLICIT MEMORIES

Explicit Memory refers to facts and experiences that one can consciously know and declare. Implicit memory

involves learning an action while the individual does not know or declare what she knows.

154

Hippocampus

Hippocampus – a neural center in the limbicsystem that processes explicit memories.

Weidenfield &

Nicolson archives

REPRESENTING IMAGES

Kosslyn has conducted a series of experiments that support the imaginal view. He found evidence that suggests that subjects are able to store detailed, accurate images in their minds.

Often, these mental maps behave in a similar way to real, visual, representations. Other data suggests that we do not store accurate images.

DUAL-TRACE THEORY

The current dialectic attempts to reconcile these findings in dual trace theory.

Dual trace theory proposes that both propositional and dimensional representations are used.

STORAGE AND FORGETTING

What represents the optimal manner in which to base your studying the material during learning?

According to the total-time hypothesis, if only one session is available for study, then it does not matter how one apportions his or her study time within the session. However if more than one study session is available then the manner in which one schedules his or her time does seem to matter.

STORAGE AND FORGETTING

Bahrick and Phelps found that subjects who studied using massed learning (cramming in one session) had inferior recall of material relative to subjects who studied under distributed learning (learning that was spread across several sessions, but equal in total time to that experienced by the massed learning subjects).

Students unfortunately often use massed practice when distributed practice would serve them much better.

STORAGE AND FORGETTING

Rehearsal has been shown to have clear effects in the encoding the memories. According to the total-time hypothesis, rehearsal’s effects are mainly due to the amount of time spent in it.

ELABORATION

However, elaboration (forging associative links among items and between the items and information already stored in long-term memory) during rehearsal does make for better retrieval.

One powerful way of accomplishing elaboration is to organize the material being encoded.

ORGANIZATION OF INFORMATION: SEMANTIC AND EPISODIC MEMORY

Semantic memory, or our general world knowledge, is common to nearly everyone, and not time-stamped in memory.

Semantic memory operates on both concepts (ideas) and schemas (cognitive networks of ideas).

ORGANIZATION OF INFORMATION: SEMANTIC AND EPISODIC MEMORY

In contrast to semantic memory, episodic memory consists of the personal, autobiographical aspects of everyone’s memory.

Episodic memories are time-tagged in that we can recall the approximate location and time of the events in our lives.

The degree to which semantic and episodic memories result from different memorial systems or processes is not entirely clear at this time.

INTERFERENCE

Both retroactive interference and proactive interference affect long-term memory:

Sometimes, older information can inhibit the acquisition of newer information. In this case proactive interference results in negative transfer.

Sometimes newer information can inhibit the acquisition of older information, retroactive interference resulting in negative transfer.

164

INTERFERENCE

Learning some new information may disruptretrieval of other information.

165

RETROACTIVE INTERFERENCE

Sleep prevents retroactive interference. Therefore, itleads to better recall.

POSITIVE TRANSFER

In positive transfer, older knowledge facilitates the acquisition of newer knowledge. For example, if one knows how to play the piano, learning to play the organ should be easier for having that knowledge.

CAPACITY

Although some researchers like Hintzman have suggested that long-term store is essentially infinite, the impossibility of designing experiments to probe the limits of the long-term store will probably prevent us from ever knowing its capacity.

It is equally difficult to design experiments to test the duration of memories in the long-term store. We cannot be certain that memories are permanent, but they have been shown to be enduring, lasting decades.

168

RETRIEVAL: GETTING INFORMATION OUT

Retrieval refers to getting information out of the memory store.

Spanky’s Yearbook Archive

Spanky’s Yearbook Archive

169

MEASURES OF MEMORY

In recognition, the person must identify an item amongst other choices. (A multiple-choice test

requires recognition.)

1. Name the capital of France.

a. Brusselsb. Romec. Londond. Paris

170

MEASURES OF MEMORY

In recall, the person must retrieve information using effort. (A fill-in-the blank test requires

recall.)

1. The capital of France is ______.

RETRIEVAL

Availability and accessibility of memory are two relevant concepts in retrieval.

Availability relates to the presence of a memory in the long-term store.

Accessibility refers to our ability to retrieve an available memory trace.

RETRIEVAL

The current controversy in psychology centers on the issue of inaccessible memories and the proposed phenomenon of repression. Some psychologists claim that victims of childhood sexual abuse repress their memories of the abuse. These memories may be released from repression into consciousness (often through the aid of a therapist). The validity of such de-repressed memories, however, is doubtful.

RETRIEVAL

The study of retrieval from LTM dates back to the work of Hermann Ebbinghaus in the late 19th century. Ebbinghaus used nonsense syllables to test his subjects retrieval because he wanted to eliminate elaboration as a confound in the study.

RETRIEVAL

This practice of using nonsense syllables of stimuli has been criticized on the grounds that the use of nonsense syllables does not rule out elaboration and that we should be looking at memory for everyday things.

In other words, memory for nonsense syllables may not resemble memory for more meaningful stimuli.

175

MEASURES OF MEMORY

In relearning, the individual shows how much time (or effort) is saved when learning material

for the second time.ListJetDaggerTreeKite…SilkFrogRing

It took 10 trialsto learn this list

ListJetDaggerTreeKite…SilkFrogRing

It took 5 trialsto learn the list

1 day laterSaving

OriginalTrials

RelearningTrials

RelearningTrials

10 510

50%

X 100

X 100

176

RETRIEVAL CUES

Memories are held in storage by a web of associations. These associations are like anchors

that help retrieve memory.

Fire Truck

truck

red

fire

heatsmoke

smellwater

hose

177

PRIMING

To retrieve a specific memory from the web of associations, you must first activate one of the strands that leads to it. This process is called

priming.

178

DÉJÀ VUDéjà Vu means “I've experienced this before.”

Cues from the current situation may unconsciously trigger retrieval of an earlier

similar experience. ©

The New

Yorker Collection, 1990. Leo Cullum from

cartoonbank.com

. All Rights Reserved

179

CONTEXT EFFECTS

After learning to move a mobile by kicking, infants most strongly respond when retested in

the same context rather than in a different context (Rovee-Collier, 1993).

Courtesy of Carolyn Rovee-Collier, Rutgers University

CUE EFFECTIVENESS AND ENCODING SPECIFICITY

Timo Mantyla found that subjects who made up their own retrieval cues were able to learn and retrieve long lists of words (500-600) with very few errors. The subjects’ self-generated cues were most effective when they were distinctive and compatible with the target word.

CUE EFFECTIVENESS AND ENCODING SPECIFICITY

Contextual cues that are present at the time of encoding may also later serve as retrieval cues. Therefore, encoding and retrieval are thought to be dependent on each other, a phenomenon Tulving and Thompson termed encoding specificity.

CUE EFFECTIVENESS AND ENCODING SPECIFICITY

Several experiments have supported the notion of encoding and specificity.

One famous experiment conducted by Godden and Baddeley showed that divers who learned lists of words either underwater or onshore are best able to recall these words when there was a match in their encoding and retrieval context (i.e., underwater or onshore). Moods and states of consciousness also have been shown to provide contextual retrieval cues

183

CONTEXT EFFECTS

Scuba divers recall more words underwater if they learned the list underwater, while they recall more

words on land if they learned that list on land (Godden & Baddeley, 1975).

Fred McConnaughey/ Photo Researchers

184

MOODS AND MEMORIESWe usually recall experiences that are consistent

with our current mood (state-dependent memory). Emotions, or moods, serve as retrieval

cues. Our memories are mood-congruent.

Jorgen Schytte/ Still Pictures

CONSTRUCTIVE NATURE OF MEMORY

Memory is not only reconstructive, or the storage of information about events as they actually happened, it is also constructive, meaning that prior experiences help subjects construct their memories, and previously stored knowledge can affect the storage of new memories. Subjects appear to be unable to distinguish between the two types.

WAR OF THE GHOSTS EXPERIMENT

In the early part of the 20th century, Sir Frederick Bartlett demonstrated the constructive nature of memory in his famous War of the Ghosts experiment.

Bartlett had British subjects read The War of the Ghosts, a Native American legend. Because the subjects were unfamiliar with many of the cultural aspects of the story, they found it difficult to read and comprehend.

WAR OF THE GHOSTS EXPERIMENT

Later, when the subjects were asked to recall the story, they distorted their memories of the story’s events to bring them more in line with their own cultural expectations.

More modern cross-cultural studies give further support of Bartlett’s view that memory is schematic and constructive.

AUTOBIOGRAPHICAL MEMORY

The memory of our own history is called autobiographical memory.

This memory is constructive and can often be quite good, but distorted, not representing what actually happened in all their details.

Rubin found that such memory is differentially good at different periods of life: middle-aged adults remember events from their childhood and the early adult years better than their more recent past.

AUTOBIOGRAPHICAL MEMORY

Typically, these memories are studied through the use of diary studies where the researcher keeps detailed autobiographies for a period of time.

Using this technique, Linton, recorded events on index cards over six years. Surprisingly, her rate of forgetting was linear, not the expected curvilinear, meaning the rate of forgetting was about the same over the entire period

AUTOBIOGRAPHICAL MEMORY

Another study by Shulster which covered a 25 year period, however, showed the more traditional serial-position effect.

191

RETRIEVAL FAILURE

Although the information is retained in the memory store, it cannot be accessed.

Tip-of-the-tongue (TOT) is a retrieval failure phenomenon. Given a cue (What makes blood cells red?) the subject says the word begins with an

H (hemoglobin).

192

MOTIVATED FORGETTING

Motivated Forgetting:People unknowingly revise their memories.

Repression: A defense mechanism that banishes anxiety-arousing thoughts, feelings, and memories from consciousness.

Sigmund Freud

Culver Pictures

193

WHY DO WE FORGET?

Forgetting can occur at any memory stage. We

filter, alter, or lose much information

during these stages.

MEMORY DISTORTIONS

According to Schacter, there are seven reasons, or sins, for why memories are distorted.1. Transience. Memory fades quickly.2. Absent-mindedness. People sometimes repeat

tasks they have just completed, or forget what they were looking for in the middle of searching.

3. Blocking. People know they have something to remember, but they can’t. An example of “tip-of-the tongue” phenomenon.

MEMORY DISTORTIONS

4. Misattribution. Not remembering where one read or heard something.

5. Suggestibility. Being susceptible to suggestion, so the memory may not be of something they actually experienced.

6. Bias. Personal biases often alter recall.7. Persistence. Inconsequential facts are

remembered as consequential facts.

SlumberTiredBedNightRestAwaken

SnoreDreamQuietPeaceRelaxedDrowsy

ON THE LIST?

Slumber Bed Sleep Drowsy Peace

SlumberTiredBedNightRestAwaken

SnoreDreamQuietPeaceRelaxedDrowsy

200

MEMORY CONSTRUCTION

While tapping our memories, we filter or fill in missing pieces of information to make our

recall more coherent.

Misinformation Effect: Incorporating misleading information into one's memory of

an event.

EYEWITNESS ACCOUNTS

Elizabeth Loftus has repeatedly demonstrated in the lab that eyewitness testimony is often less than accurate.

This poses a serious concern for our society because eyewitness testimony has been estimated to be the most important factor in many convictions.

202

Eyewitnesses reconstruct their memories when questioned about the event.

MISINFORMATION AND IMAGINATION EFFECTS

Depiction of the actual accident.

203

MISINFORMATION

Group A: How fast were the cars going when they hit each other?

Group B: How fast were the cars going when they smashed into each other?

204

MEMORY CONSTRUCTION

A week later they were asked: Was there any broken glass? Group B (smashed into) reported

more broken glass than Group A (hit).

14

32

0

10

20

30

40

50

Group A (hit) Group B (Smashed into)

Verb

Bro

ken

Gla

ss?

(%)

EYEWITNESS ACCOUNTS

In a typical procedure, Loftus showed slides to all of her subjects. The slideshow of a car accident in which a red car turned at a stop sign and runs into a pedestrian.

Then half of her subjects were asked, “Did a car pass the red car while it was stopped at the stop sign?” The other half were asked, “did another car pass the red car while it was stopped at a yield sign?”

EYEWITNESS ACCOUNTS

They were then asked to identify the original scene. Accuracy in correctly identifying the original scene was 34% higher in the group that was asked the question with the word “stop sign” in it.

Apparently, the presentation of inconsistent information to the witness has the power to make eyewitness memory less accurate.

SOURCE-MONITORING CONFUSIONS

One reason for poor eyewitness memory may be the source monitoring confusions or the ability to remember having been exposed to a stimulus, but not being able to accurately recall where such exposure occurred.

Source-monitoring confusions can lead witnesses to falsely believe that they saw or heard things that in reality were only experienced in the witness’s own mind.

208

CONSTRUCTED MEMORIES

Loftus’ research shows that if false memories (lost at the mall or drowned in a lake) are implanted in individuals, they construct

(fabricate) their memories.

Don Shrubshell

CHILDREN’S MEMORIES

Research with children has shown that children are even more susceptible to such influences than adults are, especially when leading questions are asked.

Preschool children often tell adults what the adults want to hear, include information from others when recalling the events (reality monitoring confusion), and in general make mistakes when recalling events.

210

CONSENSUS ON CHILDHOOD ABUSE

1. Injustice happens.2. Incest and other sexual abuse happen.3. People may forget.4. Recovered memories are commonplace.5. Recovered memories under hypnosis or drugs are

unreliable.6. Memories of things happening before 3 years of age

are unreliable.7. Memories, whether real or false, are emotionally

upsetting.

Leading psychological associations of the world agreeon the following concerning childhood sexual abuse:

211

STRESS HORMONES & MEMORYHeightened emotions (stress-related or

otherwise) make for stronger memories. Flashbulb memories are clear memories of emotionally significant moments or events

Scott Barbour/ Getty Im

ages

FLASHBULB MEMORIES

A recollection of an event that is so emotionally powerful that it is highly vivid and richly detailed (e.g., the assassination of President John F. Kennedy, the attack on Pearl Harbor, the attack on the Twin Towers in New York on September 11, 2001).

FLASHBULB MEMORIES